The intermediate proteasome is constitutively expressed in pancreatic beta cells and upregulated by stimulatory, low concentrations of interleukin 1 β.

A central and still open question regarding the pathogenesis of autoimmune diseases, such as type 1 diabetes, concerns the processes that underlie the generation of MHC-presented autoantigenic epitopes that become targets of autoimmune attack. Proteasomal degradation is a key step in processing of proteins for MHC class I presentation. Different types of proteasomes can be expressed in cells dictating the repertoire of peptides presented by the MHC class I complex. Of particular interest for type 1 diabetes is the proteasomal configuration of pancreatic β cells, as this might facilitate autoantigen presentation by β cells and thereby their T-cell mediated destruction. Here we investigated whether so-called inducible subunits of the proteasome are constitutively expressed in β cells, regulated by inflammatory signals and participate in the formation of active intermediate or immuno-proteasomes. We show that inducible proteasomal subunits are constitutively expressed in human and rodent islets and an insulin-secreting cell-line. Moreover, the β5i subunit is incorporated into active intermediate proteasomes that are bound to 19S or 11S regulatory particles. Finally, inducible subunit expression along with increase in total proteasome activities are further upregulated by low concentrations of IL-1β stimulating proinsulin biosynthesis. These findings suggest that the β cell proteasomal repertoire is more diverse than assumed previously and may be highly responsive to a local inflammatory islet environment.

Introduction

The proteasome is a multi-subunit complex essential for the proteolytic degradation of cellular proteins and in the generation of specific sets of bioactive peptides [1] influencing a variety of cellular processes e.g. transcriptional regulation, signaling and the regulation of the cell cycle progression [2-5].

Proteasomal activity is executed by the proteolytic core, known as the 20S proteasome. It consists of a stack of four heptameric rings: two outer α and two inner β rings [1]. The β rings are composed of catalytically active subunits (β1, β2 and β5) that cleave peptide bonds at the C-terminal side of proteins [6] with caspase-, trypsin- and chymotrypsin-like activities, respectively [7, 8]. The standard 26S proteasome contains a 19S regulatory cap that binds the polyubiquitin chain, denatures the protein, and feeds it into the proteolytic core of the proteasome [9].

Standard proteasomes (s-proteasome) assembled with β1, β2 and β5 subunits are ubiquitously expressed, but specialized proteasomes also exist and are constitutively expressed by e.g. immune cells [9], where they represent the dominant form. Formation of the proteolytic core of these specialized proteasomes involves substitution of the constitutively expressed catalytic β1, β2 and β5 subunits with the interferon (IFN)-γ-inducible β1i, β2i and β5i subunits (alternatively termed Psmb9/LMP2, Psmb10/MECL-1/LMP10 and Psmb8/LMP7, respectively) [6, 10, 11]. The immune-proteasome (i-proteasome) has an alternative 20S catalytic core where all β-subunits are replaced by IFN-γ inducible β-subunits and where the 20S-associated 19S can be replaced by the 11S (also termed PA28αβ) proteasome regulator [9, 12, 13].

When standard and inducible subunits are present in cells, the latter are preferentially incorporated into newly produced 20S proteasomes [14, 15]. Interestingly, co-expression of standard and inducible β subunits enables cells to assemble a variety of distinct 20S complexes, collectively referred to as intermediate proteasomes (int-proteasomes) [9]. The two most common int-proteasomes are composed of two inner rings containing either β1/β2/β5i or β1i/β2/β5i. These int-proteasomes are not exclusive, as other combinations have been observed, including 20S proteasome with one constitutive (β1/ β2/ β5) and one immune (β1i/ β2i/ β5i) inner ring (also called asymmetric proteasomes, [16-18]).

Immune cells permanently and many other cells under conditions of oxidative stress, inflammation, cytokine stimulation, or viral and bacterial infection express and assemble i- and int-proteasomes [9, 19]. Recently, induction of expression of such proteasomes upon exposure of human pancreatic islets and rat and mouse insulinoma cells to INFγ and β but not high concentrations of IL-1β, was reported [18, 20]. Furthermore, int-proteasomes (but not i-proteasomes) are constitutively expressed in various cells, including liver, heart, kidney, lung or colon [16, 21–24]. They constitute between 1% (heart) to 50% (liver) of the total proteasome pool [16, 21, 23, 24].

The proteasomal composition in cells has broad implications, as proteasomes exhibit diverse substrate specificities. This affects the peptide repertoire generated for presentation on major histocompatibility complex (MHC) class I molecules [13, 19, 25], signal transduction via e.g nuclear factor kappa-light-chain-enhancer of activated B cells (NFκB) [26] and protein degradation e.g. of proinsulin [27].

The s-proteasome is known to improve glucose-stimulated insulin secretion [28], regulate intracellular proinsulin levels [27] or protect against lipotoxic endoplasmic reticulum stress [29]. However, the functions of i- and int-proteasomes are poorly defined. Importantly, constitutive expression of inducible proteasome subunits in pancreatic β cells has not been described, but their induction upon INFγ and β treatment has been suggested to play a protective role against cytokine-induced apoptosis [20] and during antiviral responses [18].

Of special interest to type 1 and 2 diabetes pathogenesis is the constitutive profile of the β cell proteasomes and their regulation. Type 1 diabetes (T1D) is an autoimmune disease, in which tolerance to β cells is broken, with proinsulin serving as a major autoantigen. T1D is histologically characterized by pancreatic islet inflammation with increased levels of cytokines i.e. IL-1β, INF-γ/β and TNF-α, in the islet microenvironment [30]. Type 2 diabetes (T2D) arises when insulin secretion fails to meet demands mainly due to impaired insulin sensitivity, with β-cell oxidative and endoplasmic reticulum stress, lipotoxicity and glucotoxicity as consequences causing progressive loss of β cell functional mass [31]. All these cellular stresses induce an inflammatory response or are exacerbated by or associated with low-grade systemic inflammation via production of interleukin 1β (IL-1β) and IL-6 and recruitment and activation of innate immune cells [32, 33]. As i- and int-proteasomes can modify e.g. signal transduction and MHC I peptide presentation, their constitutive and/or induced expression in β cells by inflammatory cytokines is of high interest and therapeutic potential.

Here, we hypothesized that β cells constitutively express active non-standard proteasomes and that the expression is upregulated by innate inflammatory signals at low levels. We therefore set out to analyze the composition of proteasomes in human and mouse islets as well as in the commonly used β-cell model INS-1E cell line in non-stimulated or cytokine-stimulated conditions. We report constitutive transcription and translation of inducible proteasome subunits (β1i/ β2i/ β5i) in β-cells, albeit with lower expression levels compared to immune cell-lines. Of the inducible subunits, β5i is incorporated into active proteasomes in non-stimulated INS-1E cells, forming intermediate proteasomes that constitute 14% of total proteasomes in these cells. Furthermore, mRNA and protein expression of inducible subunits is upregulated by low concentrations of IL-1β. β5i and β1i subunits were induced in all tested cellular models while β2i was induced in mouse (but not human) islets and INS-1E cells. Consequently the composition and both constitutive and stimulated activity of proteasomes in β cells has to be considered when investigating degradation mechanisms and antigen presentation on MHC I molecules of proinsulin and other β-cell proteins.

Materials and methods

Cell culture

The rat insulinoma INS-1E cell line, a gift from Claes Wollheim and Pierre Maechler, University Medical Center, Geneva, Switzerland, was maintained as previously described [11]. The mouse insulinoma MIN6 cell line, was cultured in DMEM (Life Technologies, Nærum, Denmark) with 25 mM glucose, supplemented with 10% FBS, 0.1% Penicillin/Streptomycin (P/S), 50 uM β-mercaptoethanol and 2 mM L-glutamine. The mouse lymphocyte cell line A20, donated by Prof. Søren Buus, Department of Immunology and Microbiology, University of Copenhagen, Denmark, was cultured in RMPI-1640 (Life Technologies, Nærum, Denmark), containing 10% FBS, 1% P/S, 10 mM HEPES, 50 uM β-mercaptoethanol and 4.5 g/L D-glucose. The human T lymphocyte cell line Jurkat, also from Prof. Buus, was cultured in RPMI-1640 with 10% FBS and 1% P/S. All cells were maintained at 37o C with 5% CO2. All cell-lines were Mycoplasma negative.

Animal Care

B6 C57BL/6NRJ mice were housed, handled and sacrificed according to Danish legislation for animal experimentation and with prior approval from the local animal ethics committee, issued by the Department of Experimental Medicine, University of Copenhagen. Animal handling and procedures were conducted by researchers with FELASA certification and supervised by veterinarians.

Islet isolation and culture

Mouse islets were isolated by injection of LiberaseTM TL (Roche®, Hvidovre, Denmark) through the common bile duct to digest exocrine tissue. Islets were handpicked and either lysed immediately or cultured for 3–5 days in RPMI-1640 supplemented with 10% FBS and 1% P/S, at 37o C and 5% CO2. All data points represent separate islet collections (tested in technical triplicates) and thus denote biological variability.

Human islets were isolated from healthy, heart-beating donors by the European Consortium for Islet Transplantation (ECIT) in Milan, Italy, with local ethical approval. The obtained islets were ~ 90% pure and no apparent difference in their quality was observed. Details on islet donors are included in the Table 1. Islets were cultured as previously described in [11].

Table 1

Human islets donors information and islet preparations used during the investigation.

Islet donors information
Donor	1	2	3	4
Age	63	62	58	20
Gender (M/F)	F	F	M	M
BMI	19.5	29.3	27.8	21.8
blood group	A+	A+	O+	B+
HLA (A:B)	2,11 : 18,57	26,29 : 7,18	2,26 : 35,55	11,24 : 18,51
HLA (DR)	11,17	4,15	14,16	1,11
Cold ischemia time (h)	9	5	8.5	8
Islets culture duration (h)*	16	14	20	42
Cause of death	Cerebral bleeding	Cerebral bleeding	Cerebral bleeding	Anoxia
Source of islets	ECIT	ECIT	ECIT	ECIT
Estimated viability (%)	95	95	95	95
Estimated purity (%)	90	90	90	90
Any additional note	EBV positive			EBV positive

*time from islets isolation to shipment

Cytokine exposure

INS-1E cells were exposed to 10 ng/mL rat IFN-γ (R&D, Minnesota, USA) or 15 or150 pg/mL rat IL-1β (BD Bioscience, Lyngby, Denmark) or control medium for 24h. Human islets were exposed to 10 ng/ml human IFN-γ (BD Bioscience, New Jersey, USA) or 30 or 300 pg/ml rat IL-1β, while mouse islets were exposed to either 10 ng/mL rat IFN-γ or 50 or 300 pg/mL rat IL-1β or control medium, both for 24 hours prior to experiments.

Western blotting

Prior to experiments cells or islets were lysed in lysis buffer, consisting of 100 mM Tris (pH 8.0), 30 mM NaCl, 10 mM KCl, 10 mM MgCl2, 2% NP-40, 20 mM iodoacetamide and protease inhibitor cocktail (Life Technologies, Nærum, Denmark). Protein concentrations were measured using Bio-Rad Protein Assay Dye Reagent (Bio-Rad, Copenhagen, Denmark). Indicated amounts of proteins were loaded on Nu-Page 4–12% bis-tris gels (Thermo Fisher Scientific, Hvidovre, Denmark), and proteins were separated by SDS-PAGE. Gels were transferred to PVDF membranes using the iBLOT2 system (Thermo Fisher Scientific, Hvidovre, Denmark). Membranes were cut prior to incubation with primary antibodies (Table 2) overnight. Primary antibodies were diluted in 2% BSA in TBST (50 mM Tris pH 8, 150 mM NaCl, 0.1% Tween). Membranes were blotted with appropriate secondary antibodies for 1 hour. Blots were developed using chemiluminescence and captured using the Azure®Saphire Biomolecular Imager. Western blots were quantified using ImageJ software (v. 1.52a, [34]).

Table 2

Primary and secondary antibodies used during the investigation.

Antibody target	Company	Cat#	Dilution
β1i	Abcam	ab243556	1:10.000
β5i	Abcam	ab3329	1:5.000
β2i	Abcam	ab183506	1:1.000
Tubulin	Sigma	T6074	1:10.000
Actin	Thermofisher Scientific	MA5-11869	1:15.000
Insulin	Cell signaling	8138S	1:5.000
Anti-mouse secondary	Cell signaling	7076S	1:10.000
Anti-rabbit secondary	Cell signaling	7074S	1:10.000

Proteasome activity

INS-1E, A-20 and Jurkat cells, human and mouse islets were plated in duplicates or triplicates in 96-well plates and treated with 50 nM ONX-0914, a selective inhibitor of theβ5i subunit activity (Selleck Chemicals, Rungsted, Denmark, IC50: ~10 nM for β5i, [35]) or 2 μM MG132, a broad proteasome inhibitor (Sigma-Aldrich, Søborg, Denmark) or control medium for 2 hours prior to experiments. Chymotrypsin-, trypsin- and caspase-like activity was measured through luminescent assay using commercially available Proteasome-GloTM Assay (Promega, Nacka, Sweden) according to the manufacturer’s protocol. Depicted data are averages of either technical duplicates or triplicates as indicated. The added trypsin-like, chymotrypsin-like and caspase-like activity is referred to as total proteasome activity.

Bulk mouse islet RNA Sequencing

Five hundred mouse islets were plated and exposed to IL-1β (50 pg/mL) for 10 days or left non-exposed for 10 days. Total mRNA was extracted from the islets by employing RNeasy ® Micro Kit (Qiagen, Vedbæk, Denmark). Single-stranded, single-end sequencing libraries were generated using 35 ng of extracted RNA by means of TruSeq® Stranded mRNA Library Prep (Illumina®, Copenhagen, Denmark), and library sequencing was done with the HiSeq 4000 System (Illumina®, Copenhagen, Denmark). Sequence files were drawn to the UCSC mouse genome NCB137/mm9. Further technical and analysis details in [36] and RNA-seq raw data are accessible here: https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE110691. In brief, expression levels of all genes were estimated by Cufflink (cufflinks v2.2.1,-p 6 -G $gtf_file—max-bundle-frags 1000000000) using only the reads with exact matches. Since specific mRNA levels were analyzed no correction for multiple testing was done. Results (RPKM) for the specific genes of 3 independent experiments were analyzed by Student’s paired t-test, n = 3. RPKM for each gene is provided in Table 3.

Table 3

Low concentrations of IL-1β induce β subunit mRNA expression in mouse islets.

Gene	Ctrl 1	Ctrl 2	Ctrl 3	IL-1B1	IL-1B3	IL-1B3	Ctrlmean	IL-1β treatmentmean	P value
β5i (PSMB8)	9.38	8.54	9.03	28.13	24.57	26	8.98	26.23	0.0021
β1i (PSMB9)	6.70	5.41	6.59	13.31	16.02	15.79	6.23	15.04	0.0172
β2i (PSMB10)	12	11.75	9.57	33.99	36.21	32.18	11.11	34.13	0.001
β5(PSMB5)	34.53	37	38.27	42.61	42.78	34.31	36.6	39.9	0.4657
β1(PSMB6)	102.84	93.47	90.64	93.09	109.07	98.63	95.65	100.26	0.6015
β2(PSMB7)	79.63	76.68	79.96	85.41	83.28	72.34	78.76	80.34	0.7634

Upregulation of inducible proteasome subunits upon prolonged, low-dose exposure to IL1-β. Five hundred mouse islets were cultured and exposed to IL-1β (50 pg/mL) for 10 days. Total mRNA was extracted, and bulk (whole pancreatic islets) sequenced and genes identified using the UCSC mouse genome NCB137/mm9. mRNA levels of PSMB8, PSMB9 and PSMB10 for inducible subunits β5i, β1i and β2i, respectively, were significantly (p = <0.0005 each) upregulated by IL-1β exposure while PSMB5, PSMB6 and PSMB7 for corresponding standard proteasome subunits β5, β1 and β2 remained unchanged. Data presented as RPMK for individual islet collection (biological replicates) in the respective conditions and their means. Results were analyzed by Student’s paired t-test, n = 3.

Single-Cell RNA Sequencing of Pancreatic Islets

Each single-cell transcriptome was sequenced to ∼750,000 reads, sufficient for cell-type classification. Islet cell subpopulations were analyzed for PSMB8, PSMB9 and PSMB10 genes expression using published human islet single-cell sequencing data [37]. FastQ files were downloaded from ArrayExpress (accession: E-MTAB-5061). Data was analyzed with bcbio-nextgen (https://github.com/chapmanb/bcbio-nextgen), using the hisat2 algorithm [38] to align sequence reads to human genome version hg38 and uniquely aligned reads within RefSeq gene annotations were used to quantify gene expression with the Salmon algorithm [39]. Data is then expressed as log2 of counts per million (CPM). Only cells that passed the quality control in the original study [37] were maintained for further analysis, and the cell type classification from the original study was also maintained.

Mass spectrometry for proteasome composition analysis

INS-1E cells were grown to 90% confluence in T175 flasks. The cells were washed with HBSS before incubation with pre-warmed culture media complemented with 0.1% formaldehyde for cross-linking for 15 minutes. Next, 125 mM glycine was added for 10 minutes at 37° C to quench the formaldehyde. The advantages of live cell cross-linking vs non-crosslinking step has been evaluated in [40]. The cells were then washed three times with HBSS and centrifuged, and pellets were stored at -80°C for later proteasome composition analysis. Immuno-purification of the proteasomes from the in-vivo cross-linked lysates, was performed as previously described [41]. Briefly, proteasomes were purified by incubating the lysates with CNBr sepharose beads (GE Healthcare) covalently bound to the antibody specific for the α2 subunit of the proteasome (MCP21) (100 mg of beads for 0.8 mg antibody), using 150 million cells per 50 mg of grafted beads. The supernatant was collected, and the beads were washed three times with 40 bead volumes of washing buffer (20 mM Tris-HCl pH 7.6, 1 mM EDTA, 10% glycerol, 150 mM NaCl, 0.1% NP-40, 2 mM ATP and 5 mM MgCl2). Finally, proteins were eluted with 0.5 ml of elution buffer (20 mM Tris-HCl pH 7.6, 1 mM EDTA, 10% glycerol, 3 M NaCl, 2 mM ATP and 5 mM MgCl2). Two additional cycles of purification were conducted, reincubating the collected supernatant with antibody-grafted beads. All fractions were pooled. LC-MS/MS analysis was performed as previously described [21, 42]. Briefly, immuno-purified proteasome samples were precipitated with 20% trichloroacetic acid (TCA), washed with cold acetone and then denatured by boiling at 95°C for 30 min in the Laemmli buffer, also reversing the cross-links [41]. Proteins were alkylated and concentrated on 12% acrylamide SDS-PAGE gel as a single band, which was cut and washed. Trypsin digestion was then performed overnight at 37°C and the peptides were extracted from the gel. The digestion mixture was then dried in a Speed-Vac and resuspended with 2% acetonitrile, 0.05% trifluoroacetic acid. The peptide mixture was then analyzed by nano-LC-MS/MS using an UltiMate3000 system (Dionex) coupled to Orbitrap Fusion mass spectrometer (Thermo Fisher Scientific, Bremen, Germany). Proteins identification, validation and relative quantification were performed as previously published [41].

Statistical analysis

All samples were selected without bias and represent biological not technical variations. Distribution of islets, specifically, were randomized and independent of e.g. size and shape. As a result, samples should be homogenous and represent biological variation, and both protein expression and activity is therefore assumed to be normally distributed [43, 44]. Furthermore, normality of all expression data was tested with a Shapiro-Wilk test and found normally distributed and tested using a student t-test. Meanwhile proteasome activity and cell viability each data point is represented by a mean value of technical replicates, and as such should be normally distributed according to the central limits theorem [45]. Differences between two groups were assessed by a two-tailed Student’s t-test. All statistical analyses were done using GraphPad Prism (v. 6, La Jolla, CA). Data is represented as means ± SD or SEM. P-values of ≤0.05 were considered significant.

Results

Inducible proteasome subunits are constitutively expressed in pancreatic islets and β-cell lines

To investigate whether proteasome inducible β subunits are expressed in non-stimulated β cells we re-analyzed RNA-sequencing data of single-cells dispersed from pancreatic islets from healthy individuals [37] and found that between 3.5 to 40% of β, α and δ cells express constitutively mRNA of all inducible subunits (Fig 1A and 1B).

Fig 1

Constitutive expression of proteasome inducible subunits in islets and cell lines.

(A-B) Single cell RNA sequencing analysis of β1i, β2i and β5i gene expression in human pancreatic islet alpha, beta and delta cells from healthy individuals (n = 6). The data is shown as means with SEM. (B) presents the percentage of cells with detectable levels of inducible subunit mRNA. (C-D) SDS-PAGE and Western blot analysis of basal expression of proteasome inducible subunits in immune cell lines A20 and Jurkat, in insulinoma beta cell line INS-1E, and human islets (H-islets) and mouse islets (M-islets). Values on top of the Western blots show the amount of protein loaded. C and D are representative blots of n = 3. (E) Quantification of relative expression levels of inducible proteasome subunits normalized to tubulin (C) or actin (D) in tested cell lines/islets (n = 3, biological replicates). The data is shown as means with SD.

Next, we lysed human and mouse islets, INS-1E (β-cell insulinoma), A20 (B cell lymphoma) and Jurkat (T cell leukemia) cells and analyzed their protein contents by SDS-PAGE and Western blotting. As expected, immune cell lines (A20 and Jurkat) showed high expression of the inducible subunits (Fig 1C–1E). Interestingly, we detected relatively low but consistent expression of all three inducible subunits (β1i, β2i and β5i) in human and mouse islets and INS-1E cells (Fig 1C–1E).

INS-1E cells contain two major types of proteasomes

We next investigated the composition of proteasomes in INS-1E cells through immunoprecipitation (IP) of the 20S α2 proteasome subunit that is an obligatory member of all types of assembled proteasomes [46].

In vivo gently cross-linked INS-1E cells proteasomes were immunoprecipitated and samples analyzed by liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS). About 86% of the total proteasomes were found to be s-proteasomes with enzymatically active rings composed of β1-β2-β5 subunits (Fig 2). However, almost 14% of the precipitated proteasomes contained β5i replacing the standard β5 subunit, forming an intermediate proteasome β1-β2-β5i. We found neither β1i nor β2i in protein complexes precipitated with α2 subunit.

Fig 2

Identification of intermediate proteasomes in INS-1E cell line.

Four x108 cells were cross-linked, lysed, their proteasomes immunoprecipitated with mAb MCP21 and analyzed by LC-MS/MS. The absolute quantities of each of the six catalytic subunits measured by the LC-MS/MS method were computed to calculate the stoichiometry of 20S proteasome subtypes and the fractions of regulatory particles associated with the 20S core particle, as detailed in Experimental Procedures. INS-1E cells were cultured at standard conditions and four biological replicates were analyzed.

Finally, our MS data indicated the presence of proteasomal regulatory particles 19S and 11S (PA28αγ) within α2 proteasome complexes at the rate of 55.8% and 5.1%, respectively (Fig 2). However, with our experimental approach we cannot assign proteasome types to the detected specific regulatory particles. The remaining 39.1% of 20S corresponds to free (unactivated) proteasome [21, 41].

The β5i-selective small-molecule inhibitor ONX-0914 reduces chymotrypsin-like activity of the β cell proteasome

We next examined the profile of proteasome proteolytic activities: chymotrypsin-, trypsin- and caspase-like. The proteolytic activities were tested in unstimulated live cells by addition of specific substrates to the medium (Cell-Based Proteasome-Glo™ Assay). As shown in Fig 3A–3C intact human and mouse islets, INS-1E, MIN6 and Jurkat cells exhibited strikingly similar profiles. Under basal conditions, chymotrypsin-like activity constituted 50–60% of total proteasome activity, with the remaining 40–50% of activity almost equally divided between trypsin- and caspase-like activities. In A20 cells, trypsin- and caspase-like activities constituted a larger part of the proteasome activity than did chymotrypsin-like activity, despite the fact that they expressed the highest amounts of β5i and β1i (both subunits have chymotrypsin-like activities).

Fig 3

Total and proteolytic-specific enzymatic activities of proteasomes.

(A) Proteasome activity in human (n = 3) and mouse islets (n = 4), (B) beta cell lines: INS-1E (n = 6) and MIN6 (n = 3) and (C) immune cells: A20 and Jurkat (n = 3). All presented data points are represent biological replicates. Proteolytic-specific activities exhibited by proteasomes subunits, treated with β5i subunit specific inhibitor ONX-0914 (2 h, 50 nM) or non-specific proteasome inhibitor MG132 (2 h, 2 μM), in (D) human islets (n = 3), (E) mouse islets (n = 4) and (F) INS-1E cells (n = 4). Proteasome activity was evaluated in cultured cells/islets using Promega Proteasome Glo assay. The data is shown as luminescence per islet/cell. Statistical analysis was performed by paired t-test of treatments versus control. The data is presented as means with SD.

We next probed what portion of the observed proteasome chymotrypsin-like activity can be attributed to the inducible subunit β5i. We took an advantage of a selective and potent β5i subunit inhibitor, ONX-0914. Pretreatment of human and mouse islets and INS-1E cells with 50 nM ONX-0914 for 4 hours reduced chymotrypsin-like activity by 40%, indicating that β5i is enzymatically active in those islets and cells (Fig 3D–3F). At the same time, trypsin- and caspase-like activities were not affected by ONX-0914. As expected two μM of MG-132 treatment (broad proteasome inhibitor) almost completely blocked all three types of enzymatic activities in all tested cells and islets.

Low concentrations of IL-1β upregulate β1i, β2i and β5i subunit expression in β cells

Previous work has shown that the β1i, β2i and β5i proteasome subunits are expressed in response to IFN-γ/β in cells other than that of hematopoietic origin including β-cells [20, 21, 41] but their expression was not regulated by the high concentrations of IL-1β treatment [20]. Here we asked, if a similar to IFN-γ/β expression upregulation can be achieved by mimicking low-grade inflammation with the application of a low stimulatory concentration of IL-1β [36].

Mouse islets exposed to 50 pg/ml of IL-1β for 10 days exhibited a significantly higher mRNA expression of β5i (Psmb8), β2i (Psmb9) and β1i (Psmb10) compared to the untreated islets (Table 3). The mRNA levels for β5 (Psmb5), β2 (Psmb6) and β1 (Psmb7) genes that encode standard subunits remained unchanged after the same exposure.

Furthermore, exposure of human and mouse islets, as well as INS-1E cells, to low concentration of IL-1β for 24 h (15 pg/ml for INS-1E, 30 pg/ml for human islets and 50 pg/ml for mouse islets) induced expression of β1i, β2i and β5i (Fig 4A–4C) with the exception of β2i in human islets. The high concentration of IL-1β (150 pg/ml for INS-1E and 300 pg/ml for human/mouse islets) further induced β1i and β5i expression (but not β2i) in INS-1E cells, but failed to induce upregulation in the subunit expression in human and mouse islets (Fig 4A and 4B). As expected, low IFN-γ treatment for 24 h induced expression of all inducible subunits (Fig 4A–4C). Concentrations of IL-1β in the low range are known to increase insulin biosynthesis [47, 48]. Interestingly, induction of inducible proteasome expression by low concentrations of IL-1β or IFN-γ was associated with increased proinsulin expression levels in INS-1E cells while high concentrations of IL-1β diminished proinsulin expression (S1A and S1BFig). Furthermore, low concentration of IL-1β did not decrease the viability of mouse islets and INS1-E cells over the 24 h exposure to the cytokine (S1C and S1D Fig).

Fig 4

Cytokines induce upregulation of inducible proteasome subunits in islets/cells.

Human (A, n = 4) and mouse islets (B, n = 3) and INS-1E cells (C, n = 4) were exposed for 24 h to IL-1β at low (50 pg/ml for mouse islets and 15 pg/ml for INS-1E) or high dose (300 pg/ml for human/mouse islets and 150 pg/ml for INS-1E) or IFN-γ (10 ng/ml). Islets/cells were lysed and protein content analyzed by SDS-PAGE and Western blotting. Representative blots of four independent experiments (biological replicates) are shown (left) and quantification of inducible subunit bands relative to the tubulin (A and C) or actin (B) is presented (right). Statistical analysis was performed by paired t-tests of treatments versus control. Experiments done on individual islet donors (A and B) or biological cell replicates (C) are connected by lines.

Low concentrations of IL-1β increase proteasome activity in β cells

INS-1E cells exposed to either low concentrations of IL-1β or IFN-γ showed significant increase in all three proteasome catalytic activities. Furthermore, high concentration of IL-1β increased chymotrypsin-like and trypsin-like activity compared to controls, although the increase was less significant than that observed for low concentration exposure (Fig 5A). Mouse and human islets showed a similar pattern of increasing chymotrypsin-, trypsin- and caspase-like activity, when exposed to low concentrations of IL-1β or IFN-γ (Fig 5B and 5C). A high concentration of IL-1β did not have a significant effect on any of the proteasome-based catalytic activities in mouse and human islets.

Fig 5

Basal and cytokine induced activity of proteasome subunits in (A) INS-1E, (B) mouse and (C) human islets (all n = 3). Cells and islets were exposed to low IL-1β dose (30 pg/ml for human islets, 50 pg/ml for mouse islets and 15 pg/ml for INS-1E for 24 hours), high IL-1β dose (300 pg/ml for human islets, 300 pg/ml for mouse islets and150 pg/ml for INS-1E for 24 hours), INF-γ (10 ng/ml for human islets, 10 ng/ml for mouse islets and INS-1E for 24 hours) or control media. Statistical analysis was performed by paired t-tests of treatments versus control. Experiments done on individual islet donors (B and C) or biological cell replicates (A) are connected by lines.

Discussion

The present work shows that 1) inducible proteasome subunits are constitutively expressed in human and rodent islets and a β-cell line, 2) β5i is incorporated into an active proteasome, forming int-proteasomes and 3) inducible subunit expression is upregulated by low IL-1β concentrations. The cellular composition of proteasomes and their expressional regulation is of particular interest, because different types of proteasomes degrade proteins and peptides with different efficiency and specificity [49] influencing a variety of cellular processes including antigen presentation and thereby maintenance of peripheral tolerance or induction of autoimmunity [8, 13].

The presence of int-proteasomes as normal constituents in different tissues has been established before as they have been reported to constitute up to 50% of the total proteasome pool, depending on the tissue [16, 24], but a comprehensive investigation of proteasome composition in primary β cells or β cell models in non-stimulated conditions has not been performed. The Human Protein Atlas RNA-seq data indicated constitutive expression of the inducible β1i, β2i and β5i subunits in human pancreas, with 10% of the RNA sequencing reads originating from the islets of Langerhans and 75% coming from exocrine glandular cells [50]. Immuno-staining of the islets for specific inducible subunits detected β5i protein by one of two employed antibodies, while β1i and β2i proteins were not detected [50]. To clarify this issue, we first investigated the expression of inducible proteasome subunits in unstimulated human dispersed islet cells. Re-analysis of the previously published data set of single-cell RNA sequencing [37] uncovered substantial subpopulations of α-, β- and δ-cells that constitutively express all inducible subunits (Fig 1A and 1B) placing human islets on par with other tissues that express those subunits constitutively [9, 16]. When testing β cell models, as well as human and mouse islets, we found that all three inducible β-subunits were detectable at the protein level without the need for cytokine stimulation (Fig 1C–1E), although the expression levels of each subunit varied substantially between tested groups. β cells can therefore potentially assemble i- and int-proteasomes containing one, two or three inducible subunits without immune-stimulation. It is however important to stress that the number of β1i-positive β cells is low in human islets (Fig 1B) thus substantially limiting the possibility to assemble i-proteasome. This observation should be taken into account while investigating proteasome function and composition in pancreatic β cells.

The similar profile of expression of inducible β subunits in β, α and δ cells indicates that those subunits play parallel roles in degradation of the hormones abundantly handled by each cell type ER: insulin, glucagon and somatostatin, respectively. The cellular localization of the i- and int-proteasomes may also play a role in that process, as β5i and β1i subunits are found in close proximity to the ER while the s-proteasomes are homogenously distributed in both nucleus and cytoplasm [51].

Next, we used mass spectrometry to identify the proteasome subtypes in β cells. By immunoprecipitating the α2 subunit from INS-1E cell lysates, we purified active proteasomes. Eighty-six % of total proteasomes contained standard β subunits forming s-proteasomes (Fig 2). The remaining 14% contained only β5i subunit, whereas β1i and β2i could not be detected. Therefore, INS-1E cells constitutively express two types of proteasomes, the s-proteasome and an int-proteasome, where at least one β- ring contains a β5i subunit. We neither detected i-proteasomes nor int-proteasomes with incorporation of inducible subunits other than β5i, despite their expression in unstimulated INS-1E cells. According to the rules of cooperative assembly, β1i cannot be incorporated without β5i but the opposite is feasible [52] and thus it is theoretically possible but biologically less plausible that β1i and β2i are expressed but do not participate in the formation of a pool of active proteasomes. Alternatively, sensitivity of the antibody used for the detection of β1i may be high relative to the sensitivity of the other antibodies used for subunits detection, distorting the evaluation of the intracellular stoichiometry of the inducible subunits. The MS data that failed to detect β1i and β2i incorporation into proteasomes may thus be a more valid measure of the actual subunit stoichiometry in INS-1E cells. Furthermore, according to the rules of proteasome assembly, the lack of incorporation of β1i would prevent the incorporation of β2i subunit [52]. However, we cannot rule out that β1i and β2i are incorporated but constitute the minor portion of active proteasomes, below the detection limit of our MS method.

The enzymatically active proteasomes are generally capped on one or both ends of the central 20S proteasomal core by regulatory particles 19S or 11S, but the method employed in our study does not distinguish which type of proteasome is associated with a given regulatory particle. We have found that 56% of proteasomes in INS-1E cells contained the 19S particle known to associate with all types of proteasomes [21, 46, 53, 54], while 5.1% proteasomes contained 11S that preferentially associates with int- and i-proteasomes [12, 21, 46, 55]. This would indicate that about one third of all INS-1E int-proteasomes are bound to 11S particles while the other two third is associated with 19S particle, (hybrid int-proteasome) or not associated with any regulatory particle and thus presumably not active. Regulatory particles dictate substrate availability and specificity with 19S recognizing client proteins marked by polyubiquitin chains and 11S being involved in the degradation of short and non-ubiquitinated peptides and antigen processing for MHC I presentation [56]. As a result, their presence within assembled and active proteasomes demonstrates that unstimulated β cells contain specialized and mixed populations of proteasomes, possibly reflecting functional specificity.

When profiles of proteasome substrate-specific activities were analyzed, we found that the islets and cell lines (with the exception of A20) all showed similar proteolytic profiles. Chymotrypsin-like activity constituted between 50 and 60% of the total proteasome activity, while trypsin- and caspase-like activities were responsible for the remaining 40–50% (Fig 3A–3C). Immune cell lines have generally been reported to express a higher basal level of inducible subunits, and the i-proteasome constitutes a dominant form of their proteasomes [8]. The fact that INS-1E cells and islets share a similar proteasomal catalytic activity profile indicates that inducible subunits codetermine the activity profiles not only in immune cells.

Off note, we have observed clear differences in the proteasomal catalytic activities in two tested immune cells models, Jurkat and A20. The latter cell profile indicates persistent if not dominant incorporation of β1 subunit (with caspase-like activity) but not β1i (chymotrypsin-like activity) and diminished activity of β5 and/or β5i subunits (chymotrypsin-like activity) that are obligatory part of active proteasomes [56]. The reasons for observed differences are not known but may indicate cancer-cell-specific adjustments, human (Jurkat) vs mouse (A20) divergence or reflect more physiologically important differences between T (Jurkat) and B (A20) cells. Finally, it is plausible that proteasome activity in A20 cells is additionally modified by e.g. post-translational modifications or altered transcription of proteasomal activators, as reviewed in [57].

We next pretreated islets and cells with a β5i selective small-molecule inhibitor, ONX-0194, and found a 30 to 50% reduction in chymotrypsin-like activity, further indicating that the β5i subunit is proteolytically active in β cells (Fig 3D–3F).

Interpretation of the pathophysiological consequences of the proteasomes diversity in β cells requires better understanding of factors influencing its expression and composition. The human genes coding for β5i and β1i map to chromosome 6 precisely between the DNA sequences coding for human leucocyte antigen (HLA)-DQ, HLA-DM and Transporter 1 ATP Binding Cassette Subfamily B Member (TAP) 1 and 2 (S2 Fig), genes known to be major determinants of antigen presentation and predisposing to autoimmune diseases, including type 1 diabetes [58]. The promoter region of β5i contains binding sites for the NFκB transcription factor (S2 Fig), but high concentrations of IL-1β, a strong inducer of NFκB [59], do not increase β5i expression [20]. Accordingly, studies in neurons have shown that high concentration of IL-1β induces Early growth response 1 protein (Egr1) that strongly inhibits transcriptional activity at the β5i promoter [60]. At the same time, Freudenburg et al. speculated that since viral infections induce IL-1β synthesis, iNOS expression and nitric oxide production impeding on mitochondrial function, the resulting reduction in ATP levels would trigger i-proteasome activation and generation of altered peptides that may be immunogenic and enable killing of infected target cells as an appropriate host antiviral response [18]. How can these apparently disparate IL-1β functions be reconciled? We suggest that cytokine concentration and/or duration of exposure are a key determinants of cell fate. It has been reported that low IL-1β concentrations (0.01–0.1 ng/ml) are stimulating and protective for β cells e.g. they improve insulin biosynthesis and secretion and increase β cell proliferation, while higher concentrations (5–20 ng/ml) can induce cell apoptosis and necrosis through e.g. induction of endoplasmic reticulum and mitochondrial stress [47, 48]. These two outcomes employ different cellular pathways, the stimulatory pathway depending on PKC and phospholipases and the toxic pathway on NFκB signaling. We therefore used 10–100 fold lower IL-1β concentrations in our experiments compared to previous publications [20] as well as, in case of mouse islet used for bulk sequencing (Table 3), we extended the islet exposure to IL-1β up to 10 days in order to better mimic long-term low-grade inflammation. Indeed, we found that treatment of human and mouse islets and INS-1E cells with low concentrations of IL-1β increased the mRNA (Table 3) and protein expression of all inducible proteasome β subunits (except of β2i in human islets, Fig 4), while it had no impact on mRNA levels of standard subunits (Table 3) and increased all substrate-specific proteolytic activities in human and mouse islets and INS-1E cells (Fig 5). Off note, 10 day mouse islets exposure to low concentrations of IL-1β did diminished their glucose induced insulin secretion but did not reduced islets insulin content nor induced endoplasmic reticulum stress or cell death as reported in Ibarra et al. Mol Cell Endocrinol. 2019.

Our results indicate that cytokine concentration is critical when evaluating the regulatory role of cytokines in proteasome expression and activity.

Proteasomes process proteins of both endogenous and exogenous origin and produce peptides that are complexed with MHC I. The shift in composition of proteasomes towards i-proteasomes, changes the peptide repertoire from non-immunogenic to immunogenic [18, 61] and can contribute to the progression towards autoimmune diabetes [18, 20, 62, 63]. Importantly, the observed differences in cytokine action may reflect changing conditions in the islet microenvironment during inflammatory or metabolic stress. IL-1β is a central promoter of low-grade inflammation and protection against certain viral infections, including influenza [64]. One of the possible host protective mechanisms engaged by this cytokine could involve expression and assembly of int- and i-proteasomes that would result in an increased presentation of viral antigens and/or modified self-antigens, thereby enabling T effector-cell dependent eradication of infected cells. Similarly, low grade inflammation and local IL-1β production in the islet microenvironment, could facilitate neoepitope presentation by β cells through preferential incorporation of inducible subunits to form int- or i-proteasomes. Interestingly, cells deficient in β5i show lower MHC I expression and peptide presentation, and β5i pharmacological inhibition slows disease progression in mouse models of inflammatory diseases such as arthritis and lupus [35, 65]. Furthermore, β5i has also been implicated in type 1 diabetes, and its inhibition has been shown to have a protective effect [35]. This could reflect the fact that the MHC I peptides repertoire is at least in part dependent upon the activity of β5i subunit in β cells expressing int- or i-proteasomes.

The role of immune- and, especially, intermediate proteasomes in β cell pathophysiology remains to be uncovered in detail. The perspective that differential proteasome subunit expression dictates the repertoire of β-cell neoepitopes presented by MHC I deserves future investigation. Discoveries in this field could lead to targeted proteasome inhibition as treatment options in diseases with an autoimmune component. In this study, we lay the groundwork for such future investigations. For the first time, we show that int-proteasomes are constitutively expressed and active in β cells and that inducible proteasome subunits can be upregulated in β cells in response to stimulatory low concentrations of IL-1β along with increases in total proteasome activities.

Cytokine induced increase in proinsulin levels in INS-1E cells.

Lysates of cells exposed to IL-1β at low (15 pg/ml) and high concentration (150 pg/ml), IFN-γ (10ng/ml) and control medium, were run on SDS-PAGE and subjected to Western blotting in A) and proinsulin band intensity normalized to tubulin in B). C) and D) Mouse islets and INS1-E cell viability was tested. Staining reagent (AlamarBlue) was added to the cell culture for 4 h, incubated at 37oC, and the resulting fluorescence was read on a plate reader. Statistical analysis was performed by paired t-tests of treatments versus control. The data is shown as means with SD.

(TIF)

Click here for additional data file.

Genetic localization of β5i and β1i genes.

Genes for β5i (PSMB8) and β1i (PSMB9) reside in the MHC-II region on human chromosome 6. Presented transcription factors were included in UCSC genome browser genome GRCh37 and visualized with integrated regulation from ENCODE-track option. For clarity and in relevance to the current publication, only some transcription factors are presented.

(TIF)

Click here for additional data file.

6 Nov 2019

PONE-D-19-24067

The intermediate proteasome is constitutively expressed in pancreatic beta cells and upregulated by stimulatory, non-toxic concentrations of interleukin 1 b

PLOS ONE

Dear Prof. Marzec,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

==============================

As pointed out by the reviewers, a number of issues could be addressed or at least clarified in the text.

==============================

We would appreciate receiving your revised manuscript by Dec 16 2019 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Corentin Cras-Méneur, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

1. When submitting your revision, we need you to address these additional requirements. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. PLOS ONE now requires that authors provide the original uncropped and unadjusted images underlying all blot or gel results reported in a submission’s figures or Supporting Information files. This policy and the journal’s other requirements for blot/gel reporting and figure preparation are described in detail at https://journals.plos.org/plosone/s/figures#loc-blot-and-gel-reporting-requirements and https://journals.plos.org/plosone/s/figures#loc-preparing-figures-from-image-files. When you submit your revised manuscript, please ensure that your figures adhere fully to these guidelines and provide the original underlying images for all blot or gel data reported in your submission. See the following link for instructions on providing the original image data: https://journals.plos.org/plosone/s/figures#loc-original-images-for-blots-and-gels.

In your cover letter, please note whether your blot/gel image data are in Supporting Information or posted at a public data repository, provide the repository URL if relevant, and provide specific details as to which raw blot/gel images, if any, are not available. Email us at plosone@plos.org if you have any questions.

3. Please amend the manuscript submission data (via Edit Submission) to include author Nils Billestrup.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This is a descriptive study that examines the expression of subunits of the immune proteasome in pancreatic islets. The authors have used previously published RNAseq evidence to suggest that components of the immune proteasome are expressed at the mRNA level in b-cells and other islet endocrine cells. Overall the study is consistent with other reports showing immune proteasome expression under various conditions with new information suggestive that long exposures with low concentrations of IL-1 are capable of stimulating subunit expression. There are a number of concerns that could be addressed or clarified in the text. Using RNAseq the authors provide in support of immune proteasome expressed in b-cells under basal conditions. How was RNAseq used to quantify mRNA levels? There is also some concern with the MS analysis of cross-linked proteasomes via immunoprecipitation approaches. It is not clear how the crosslinking on intact cells crosslinks the proteasome found in the cytoplasm (Fig 2). Some controls could be helpful with this analysis to support the authors conclusions.

There is also concern regarding “non-toxic” concentrations of IL-1. It is not clear that mouse or human islets are sensitive to IL-1 alone and therefor all concentrations of IL-1 are non-toxic. It is suggested that the text be modified to low and high concentrations to be more consistent with the biological action. Also- the merits of a 10 day exposure are not clear- what is the time- and concentration-dependence of IL-1 on immune proteasome subunit expression?

Quantification of RNAseq? How is it possible to determine a p-value on these studies (Table 3) and how was quantification performed. Number of cells expressing the targets or the level of expression (i.e. number of reads?). Further, was this confirmed by PCR? Also, was this one isolation of islets and the n represent the number of cells sequenced, or were multiple isolations performed for each single cell sequence?

Reviewer #2: The authors describe the constitutive expression of intermediate proteasomes in pancreatic beta cells. This expression is upregulated by interleukin 1b (IL1b). Mining of available RNA-seq data of single cells from pancreatic islets of healthy individuals and analysis of human and mouse islets, INS-1E (insulinoma), A20 (lymphoma) and Jurkat (leukemic cells) cells, revealed high expression of the inducible proteasome subunits in immune cells and low but consistent expression in the pancreatic cells. By LC-MS/MS analysis, proteasomes from INS-1E cells were classified either as standard (s-proteasomes) or beta5i-containing (14%) but no proteasomes were detected with the beta1i and beta2i subunits. With the exception of the A20 cells, all other cells types tested had proteasomes exhibiting mainly chymotrypsin-like activity (50-60%) followed by trypsin- and caspase-like activity. Incubation of human/mouse islets with ONX-0914 – an inhibitor of beta5i – led to reduction in chymotrypsin-like activity. Moreover, IL-1b upregulated the expression of the inducible subunits beta-1 (b1i), -2 (b2i) and -5 (b5i) and the proteasome activity in beta cells.

The findings are important as our knowledge of the proteasome activities in beta cells is still fairly limited. The authors provide a comprehensive picture of proteasome composition, especially with respect to the inducible subunits, in beta cells. Overall, the study is technically sound and the results support the conclusions. Statistical analysis of the data has been performed. I only have a few minor comments:

1. The authors should address the discrepancy in the level of chymotrypsin-like activity in the proteasomes of beta cells and Jurkat cells vs. A20 cells (lines 305-307). This is only mentioned in passing in the discussion (line 452).

2. Reporting the viability of cells (e.g. in figure 5) would be helpful to better assess the reported effects and the effect of treatment on the cells or islets.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

26 Nov 2019

PONE-D-19-24067

The intermediate proteasome is constitutively expressed in pancreatic beta cells and upregulated by stimulatory, non-toxic concentrations of interleukin 1 b

Response to Reviewers:

Reviewer #1: This is a descriptive study that examines the expression of subunits of the immune proteasome in pancreatic islets. The authors have used previously published RNAseq evidence to suggest that components of the immune proteasome are expressed at the mRNA level in b-cells and other islet endocrine cells. Overall the study is consistent with other reports showing immune proteasome expression under various conditions with new information suggestive that long exposures with low concentrations of IL-1 are capable of stimulating subunit expression.

There are a number of concerns that could be addressed or clarified in the text. Using RNAseq the authors provide in support of immune proteasome expressed in b-cells under basal conditions. How was RNAseq used to quantify mRNA levels?

Two types of RNAseq data are included in the manuscript. Fig.1A presents human single cell sequencing data and Table 3 presents bulk mouse pancreatic islet RNA sequencing data.

Quantification of single-cell RNA sequencing of human pancreatic islets was performed through data analysis with bcbio-nextgen toolkit (https://bcbio-nextgen.readthedocs.io/en/latest/), followed by the Salmon algorithm (Patro, Duggal et al. 2017) for quantitation of mRNA counts. Data is then expressed as log2 of counts per million (CPM). The original use of data and more analysis details are included in, cited in the Material and Methods section, reference (Segerstolpe, Palasantza et al. 2016). Material and Method section has been amended to reflect changes.

Table 3 and its description (as well as manuscript text at the lane 203-216; all lane references refer to marked copy of the manuscript) has been modified to underlie the fact that data comes from bulk mouse islet sequencing (see below).

Quantification of RNAseq? How is it possible to determine a p-value on these studies (Table 3) and how was quantification performed. Number of cells expressing the targets or the level of expression (i.e. number of reads?). Further, was this confirmed by PCR? Also, was this one isolation of islets and the n represent the number of cells sequenced, or were multiple isolations performed for each single cell sequence?

We thank the Reviewer for raising this issue as it made us realize that the data should be reported in a different and more detailed way. Currently the data reports on in RPKM for the specific genes of 3 independent experiments as well as their means and is analyzed by Student’s paired t-test. The conclusions remain the same. As mentioned in the material and methods section Table 3 reports on whole islets bulk RNAseq and was obtained from 3 separate isolations of islets from several mice per condition. The method of quantification is described in detail and a reference to the original publication is provided. The expression of six genes reported in this study have not been validated by PCR but multiple other genes in the original study were for ex. insulin, Mafa, Pdx-1 and Ucn3.

Text changes can be found in lanes 203-216, 354-355 and overhauled Table 3 at lane 346.

There is also some concern with the MS analysis of cross-linked proteasomes via immunoprecipitation approaches. It is not clear how the crosslinking on intact cells crosslinks the proteasome found in the cytoplasm (Fig 2). Some controls could be helpful with this analysis to support the authors conclusions.

Co-author Marie-Pierre Bousquet’s group has set the original protocol to in vivo crosslink proteins on intact cells of various origins and then to very efficiently immmunopurify proteasome complexes after cell lysis. They have shown that this crosslinking strategy helps stabilizing Proteasome Interacting Proteins (PIPs), in particular important proteasome regulators, like the 19S and PA28 activators, and enables to characterize the whole diversity of proteasome complexes. In their 1st publication (Bousquet-Dubouch MP et al., Mol. Cell. Proteomics, 2009), the comparison of crosslinked cells and non-crosslinked cells (used as a control) has clearly demonstrated the benefit of formaldehyde crosslinking step on the efficacy of recovery of proteasome subunits and PIPs. Since then, the protocol has been successfully applied to study proteasome complexes in different subcellular compartments in leukemic cells (Fabre B. et al, Mol. Cell. Proteomics, 2013) and in cells and tissues of various origins (Fabre B. et al, J. Proteome Res., 2014; Fabre B. et al, Mol. Sys. Biol., 2015).

Material and method section has been amended to include the original publication for the protocol reference (lane 234).

There is also concern regarding “non-toxic” concentrations of IL-1. It is not clear that mouse or human islets are sensitive to IL-1 alone and therefor all concentrations of IL-1 are non-toxic. It is suggested that the text be modified to low and high concentrations to be more consistent with the biological action.

We agree with the reviewer and have corrected the text accordingly.

Also- the merits of a 10 day exposure are not clear- what is the time- and concentration-dependence of IL-1 on immune proteasome subunit expression?

We thank the Reviewer for raising this issue. The rational for doing the 10 days exposure to low doses of IL-1 was to mimic and better resemble the low-grade inflammation observed in diabetes. Although, even 24h exposure to IL-1 induces significant changes in proteasome subunit expression and activity the longer exposure, more characteristic for diabetes pathology could result in a loss of inducible subunits expression and induction of other detrimental to beta cells processes. However as reported in our collaborators publication (Ibarra et al. Mol Cell Endocrinol. 2019) prolonged exposure of mouse islets to IL-1 resulted in decrease of glucose stimulated insulin secretion but did not reduce cellular insulin content, did not induce endoplasmic reticulum stress or cell death and inducible subunits expression was maintained. This experiment provides a time perspective underlying the relevance of the observed short- and long-term subunits induction to the pathology of diabetes. We have modified the manuscript text at lane 502-511.

Reviewer #2: The authors describe the constitutive expression of intermediate proteasomes in pancreatic beta cells. This expression is upregulated by interleukin 1b (IL1b). Mining of available RNA-seq data of single cells from pancreatic islets of healthy individuals and analysis of human and mouse islets, INS-1E (insulinoma), A20 (lymphoma) and Jurkat (leukemic cells) cells, revealed high expression of the inducible proteasome subunits in immune cells and low but consistent expression in the pancreatic cells. By LC-MS/MS analysis, proteasomes from INS-1E cells were classified either as standard (s-proteasomes) or beta5i-containing (14%) but no proteasomes were detected with the beta1i and beta2i subunits. With the exception of the A20 cells, all other cells types tested had proteasomes exhibiting mainly chymotrypsin-like activity (50-60%) followed by trypsin- and caspase-like activity. Incubation of human/mouse islets with ONX-0914 – an inhibitor of beta5i – led to reduction in chymotrypsin-like activity. Moreover, IL-1b upregulated the expression of the inducible subunits beta-1 (b1i), -2 (b2i) and -5 (b5i) and the proteasome activity in beta cells.

The findings are important as our knowledge of the proteasome activities in beta cells is still fairly limited. The authors provide a comprehensive picture of proteasome composition, especially with respect to the inducible subunits, in beta cells. Overall, the study is technically sound and the results support the conclusions. Statistical analysis of the data has been performed. I only have a few minor comments:

1. The authors should address the discrepancy in the level of chymotrypsin-like activity in the proteasomes of beta cells and Jurkat cells vs. A20 cells (lines 305-307). This is only mentioned in passing in the discussion (line 452).

We have made changes to the manuscript to clearly underlie observed discrepancies and pointed to their possible sources although we have to admit that at this stage we do not understand the observed proteasome profile in A20 cells (lane 466-477). The proteasome activity and/or substrate specific activity can be regulated on multiple levels and the issue has been recently reviewed in Kors et al. Front. Mol. Biosci., 16 July 2019, https://doi.org/10.3389/fmolb.2019.00048, although, again, it does not provide a clear explanation of our observation.

2. Reporting the viability of cells (e.g. in figure 5) would be helpful to better assess the reported effects and the effect of treatment on the cells or islets.

We agree with the Reviewer and have added data on mouse islets and INS1-E cell viability to the S1 figure and made appropriate changes in the manuscript text (lane 367-368) adding also references from previously published work e.g. Ibarra et al. Mol Cell Endocrinol. 2019 PMID: 31362031 Fig. 1E.

References

Patro, R., G. Duggal, M. I. Love, R. A. Irizarry and C. Kingsford (2017). "Salmon provides fast and bias-aware quantification of transcript expression." Nat Methods 14(4): 417-419.

Segerstolpe, A., A. Palasantza, P. Eliasson, E. M. Andersson, A. C. Andreasson, X. Sun, S. Picelli, A. Sabirsh, M. Clausen, M. K. Bjursell, D. M. Smith, M. Kasper, C. Ammala and R. Sandberg (2016). "Single-Cell Transcriptome Profiling of Human Pancreatic Islets in Health and Type 2 Diabetes." Cell Metab 24(4): 593-607.

Submitted filename: Response to Reviewers Khilji et al.docx

Click here for additional data file.

2 Jan 2020

PONE-D-19-24067R1

The intermediate proteasome is constitutively expressed in pancreatic beta cells and upregulated by stimulatory, low concentrations of interleukin 1 b

PLOS ONE

Dear Prof. Marzec,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

While most of the comments in the manuscript have been addressed according to the comments made by the reviewers, two points remain to be clarified in the text for the manuscript to be accepted.

We would appreciate receiving your revised manuscript by Feb 16 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Corentin Cras-Méneur, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (if provided):

The revised version of the manuscript addresses most of the comments raised by the original reviewers.

There are only two points that remained to be clarified in the manuscript:

• The authors need to mention and justify that Normal distribution needs too be assumed for the use of parametric tests.

• The authors need to emphasize and discuss in the text that the replicates are technical replicates (3 isolations) and not biological replicates.

[Note: HTML markup is below. Please do not edit.]

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

21 Jan 2020

PONE-D-19-24067

The intermediate proteasome is constitutively expressed in pancreatic beta cells and upregulated by stimulatory, non-toxic concentrations of interleukin 1 b

Response to Editor Comments

The revised version of the manuscript addresses most of the comments raised by the original reviewers.

There are only two points that remained to be clarified in the manuscript:

• The authors need to mention and justify that Normal distribution needs too be assumed for the use of parametric tests.

The following text has been added to Statistical analysis section of the manuscript:

All samples were selected without bias and represent biological not technical variations. Distribution of islets, specifically, were randomized and independent of e.g. size and shape. As a result, samples should be homogenous and represent biological variation, and both protein expression and activity is therefore assumed to be normally distributed (43, 44). Furthermore, normality of all expression data was tested with a Shapiro-Wilk test and found normally distributed and tested using a student t-test. Meanwhile proteasome activity and cell viability data is represented by a mean value of technical replicates, and as such should be normally distributed according to the central limits theorem (45).

• The authors need to emphasize and discuss in the text that the replicates are technical replicates (3 isolations) and not biological replicates.

The replicates presented in all figures, irrespective if those are cells or pancreatic islets, are biological not technical replicates.

Islet isolation: islets were isolated on separate days and from separate group of mice and that constitutes biological replications. Similarly, in case of human islets obtained from deceased donors, isolated islets represent biological variability (thus are replications) not a technical replicate.

Proteasome studies: for each data point cells were separately cultured (minimum two weeks), plated (on different days) and collected. That procedure constitutes biological replication for cell culture experiments. Then each data set was run, in minimum, as three technical replicates and the mean of those is presented.

We have amended the manuscript text to emphasize the fact that presented data is based on biological replicates. See lanes: 164-164, 287, 311 (already mentioned), 334-335, 361-362, 383, 403 (already mentioned).

Submitted filename: Response to Reviewers Khilji et al Jan 21 2020.docx

Click here for additional data file.

22 Jan 2020

The intermediate proteasome is constitutively expressed in pancreatic beta cells and upregulated by stimulatory, low concentrations of interleukin 1 b

PONE-D-19-24067R2

Dear Dr. Marzec,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Corentin Cras-Méneur, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

5 Feb 2020

PONE-D-19-24067R2

The intermediate proteasome is constitutively expressed in pancreatic beta cells and upregulated by stimulatory, low concentrations of interleukin 1 β

Dear Dr. Marzec:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Corentin Cras-Méneur

Academic Editor

PLOS ONE